This project is based on the hypothesis that neural mechanisms in the control of respiration are more important than traditionally believed. The studies use the neural output of the respiratory control system rather than ventilation, allowing use of "open-loop" experimental conditions. The input-output characteristics of the control system to neural stimuli are being studied. It is proposed to extend studies on the ability of the medullary-positive controller to maintain its output as a feedback-stimulation. Current work suggests that this mechanism is located in a network of neurons separate from the respiratory output neurons and those involving phase switching. It is believed that this network may show tonic activity, but its anatomical substrate is unknown. Unit recording of neurons in the medullar and pons, particularly in the reticular formation will be carried out in an attempt to characterize activity which may represent this function.